Brain-inspired global-local learning incorporated with neuromorphic computing

There are two principle approaches for learning in artificial intelligence: error-driven global learning and neuroscience-oriented local learning. Integrating them into one network may provide complementary learning capabilities for versatile learning scenarios. At the same time, neuromorphic computing holds great promise, but still needs plenty of useful algorithms and algorithm-hardware co-designs to fully exploit its advantages. Here, we present a neuromorphic global-local synergic learning model by introducing a brain-inspired meta-learning paradigm and a differentiable spiking model incorporating neuronal dynamics and synaptic plasticity. It can meta-learn local plasticity and receive top-down supervision information for multiscale learning. We demonstrate the advantages of this model in multiple different tasks, including few-shot learning, continual learning, and fault-tolerance learning in neuromorphic vision sensors. It achieves significantly higher performance than single-learning methods. We further implement the model in the Tianjic neuromorphic platform by exploiting algorithm-hardware co-designs and prove that the model can fully utilize neuromorphic many-core architecture to develop hybrid computation paradigm.

The Energy Homeostasis Principle: A Naturalistic Approach to Explain the Emergence of Behavior

It is still elusive to explain the emergence of behavior and understanding based on its neural mechanisms. One renowned proposal is the Free Energy Principle (FEP), which uses an information-theoretic framework derived from thermodynamic considerations to describe how behavior and understanding emerge. FEP starts from a whole-organism approach, based on mental states and phenomena, mapping them into the neuronal substrate. An alternative approach, the Energy Homeostasis Principle (EHP), initiates a similar explanatory effort but starts from single-neuron phenomena and builds up to whole-organism behavior and understanding. In this work, we further develop the EHP as a distinct but complementary vision to FEP and try to explain how behavior and understanding would emerge from the local requirements of the neurons. Based on EHP and a strict naturalist approach that sees living beings as physical and deterministic systems, we explain scenarios where learning would emerge without the need for volition or goals. Given these starting points, we state several considerations of how we see the nervous system, particularly the role of the function, purpose, and conception of goal-oriented behavior. We problematize these conceptions, giving an alternative teleology-free framework in which behavior and, ultimately, understanding would still emerge. We reinterpret neural processing by explaining basic learning scenarios up to simple anticipatory behavior. Finally, we end the article with an evolutionary perspective of how this non-goal-oriented behavior appeared. We acknowledge that our proposal, in its current form, is still far from explaining the emergence of understanding. Nonetheless, we set the ground for an alternative neuron-based framework to ultimately explain understanding.

Mastery of Technological Pedagogical And Content Knowledge (TPACK) Prospective Teacher In ICT Expertise Brawijaya University

The Information Technology Education (ITE) study program, Faculty of Computer Science, Universitas Brawijaya requires its students to take part in Pengenalan Lapangan Persekolahan (PLP) 2 according to Permenristekdikti No. 55 of 2017 in order to be able to produce prospective teachers who have the competence of educators. This study describes the gap in mastery of competencies with the TPACK framework based on the results of PLP 2 activities which are compared with standard values using a discrepancy evaluation model. The results of the gap are mapped using the method Importance Performance Analysis (IPA) to determine the priority scale for improvement of variables according to positions in certain quadrants. Through the IPA method, the variables that are prioritized to improve their mastery are TPK and PCK because they have very small gaps. Recommendations are given for the TPK variable to be given a pretest and posttest on the material for preparing teaching tools and training in the preparation of lesson plans. For the PCK variable, should be given pretest and posttest to the study material for theoretical and practical learning scenarios, the activities are needed lesson study which is carried out at least twice.

Influence of COVID-19 Pandemic in Students’ Distribution of Time in Distant Learning

Statistical analysis offers unprecedented opportunities to identify learning strategies. This fact has been boosted in the COVID-19 pandemic because of the data obtained in distant learning scenarios due to the confinement. This article deals with the identification of students’ strategies in different distant learning scenarios such as working autonomously as a support for face-to-face classes or learning autonomously in COVID-19 confinement. We have measured the influence of parameters such as the time they spent in self-evaluation, the scores obtained through this process and the distribution of time when studying autonomously. We have only detected significant results that guide to a better learning strategy when we include time parameters, such as the time between studying sessions or the time between students’ first session and their final exam. We demonstrate that students that started to study earlier and more dispersed get better success ratio (not necessarily better scores) than those that started later and do it more concentrate. The findings from this study suggest that the same amount of time spent in autonomous learning optimizes its effectiveness when it is extended in time. This learning strategy was found more often in COVID-19 confinement, where students were forced to stay at home.

MANAGEMENT OF ICT-MEDIATED LEARNING: A CURRICULUM DESIGN PROPOSAL FROM UNIVERSIDAD NACIONAL DE CHIMBORAZO.

The current situation in the field of education demands teachers who are capable of functioning in new learning scenarios where the possibilities offered by ICT for information acquisition and communication processes are enormous. In this sense, it is necessary to have postgraduate programs that contribute to the development of digital skills in teachers. The main purpose of this work is to propose the curricular design for a Master's program in Education, Mention in Management of Learning Mediated by ICT, offered by Universidad Nacional de Chimborazo in Ecuador. For this, a qualitative research was undertaken in order to characterize and determine the most important features of each module of the curriculum. A documentary research design was applied through the PICOC method (Population, Intervention, Comparison, Outcome, Context). The result of this work was a curricular mesh that consists of 12 study modules wherein aspects such as: digital literacy for the new society were addressed; didactics in new digital environments; the design and development of content and digital resources for learning; new ways of learning and innovating in education; as well as research in educational technology.

Evaluating Federated Learning Scenarios in a Tumor Classification Application

Federated Learning is a new area of distributed Machine Learning (ML) that emerged to deal with data privacy concerns. In this approach, each client has access to a local and private dataset. They only exchange the model weights and updates. This paper presents a Federated Learning (FL) approach to a cloud Tumor-Infiltrating Lymphocytes (TIL) application. The results show that the FL approach outperformed the centralized one in all evaluated ML metrics. It also reduced the execution time although the financial cost has increased.

A Visual Mining Approach to Improved Multiple-Instance Learning

Multiple-instance learning (MIL) is a paradigm of machine learning that aims to classify a set (bag) of objects (instances), assigning labels only to the bags. This problem is often addressed by selecting an instance to represent each bag, transforming an MIL problem into standard supervised learning. Visualization can be a useful tool to assess learning scenarios by incorporating the users’ knowledge into the classification process. Considering that multiple-instance learning is a paradigm that cannot be handled by current visualization techniques, we propose a multiscale tree-based visualization called MILTree to support MIL problems. The first level of the tree represents the bags, and the second level represents the instances belonging to each bag, allowing users to understand the MIL datasets in an intuitive way. In addition, we propose two new instance selection methods for MIL, which help users improve the model even further. Our methods can handle both binary and multiclass scenarios. In our experiments, SVM was used to build the classifiers. With support of the MILTree layout, the initial classification model was updated by changing the training set, which is composed of the prototype instances. Experimental results validate the effectiveness of our approach, showing that visual mining by MILTree can support exploring and improving models in MIL scenarios and that our instance selection methods outperform the currently available alternatives in most cases.

Zen and the art of model adaptation: Low-utility-cost attack mitigations in collaborative machine learning

In this study, we aim to bridge the gap between the theoretical understanding of attacks against collaborative machine learning workflows and their practical ramifications by considering the effects of model architecture, learning setting and hyperparameters on the resilience against attacks. We refer to such mitigations as model adaptation. Through extensive experimentation on both, benchmark and real-life datasets, we establish a more practical threat model for collaborative learning scenarios. In particular, we evaluate the impact of model adaptation by implementing a range of attacks belonging to the broader categories of model inversion and membership inference. Our experiments yield two noteworthy outcomes: they demonstrate the difficulty of actually conducting successful attacks under realistic settings when model adaptation is employed and they highlight the challenge inherent in successfully combining model adaptation and formal privacy-preserving techniques to retain the optimal balance between model utility and attack resilience.